Pentax/Camera Marketplace |
Pentax Items for Sale |
Wanted Pentax Items |
Pentax Deals |
Deal Finder & Price Alerts |
Price Watch Forum |
My Marketplace Activity |
List a New Item |
Get seller access! |
Pentax Stores |
Pentax Retailer Map |
Pentax Photos |
Sample Photo Search |
Recent Photo Mosaic |
Today's Photos |
Free Photo Storage |
Member Photo Albums |
User Photo Gallery |
Exclusive Gallery |
Photo Community |
Photo Sharing Forum |
Critique Forum |
Official Photo Contests |
World Pentax Day Gallery |
World Pentax Day Photo Map |
Pentax Resources |
Articles and Tutorials |
Member-Submitted Articles |
Recommended Gear |
Firmware Update Guide |
Firmware Updates |
Pentax News |
Pentax Lens Databases |
Pentax Lens Reviews |
Pentax Lens Search |
Third-Party Lens Reviews |
Lens Compatibility |
Pentax Serial Number Database |
In-Depth Reviews |
SLR Lens Forum |
Sample Photo Archive |
Forum Discussions |
New Posts |
Today's Threads |
Photo Threads |
Recent Photo Mosaic |
Recent Updates |
Today's Photos |
Quick Searches |
Unanswered Threads |
Recently Liked Posts |
Forum RSS Feed |
Go to Page... |
|
Search this Thread |
12-08-2020, 05:01 AM - 1 Like | #1 |
Use of Astrotracer with K-70
Hi there I am very new to field of astrophotography and have been playing around with my new K-70 camera and the GPS unit. From what I have read, the Astrotracer is meant to be used in BULB mode not USER MODE 3 (ASTROPHOTO). Is that correct? However, I have also read that the ASTROPHOTO MODE is compatible with the Astrotracer. This has confused me immensely. When should the ASTROPHOTO MODE be used? Thanks in anticipation Doug | |
These users Like Simmo25's post: |
12-08-2020, 05:08 AM | #2 |
Hi there I am very new to field of astrophotography and have been playing around with my new K-70 camera and the GPS unit. From what I have read, the Astrotracer is meant to be used in BULB mode not USER MODE 3 (ASTROPHOTO). Is that correct? However, I have also read that the ASTROPHOTO MODE is compatible with the Astrotracer. This has confused me immensely. When should the ASTROPHOTO MODE be used? Thanks in anticipation Doug Cheers! | |
12-08-2020, 06:21 AM | #3 |
I don't own a K-70 or a GPS unit so I am merely guessing based on the names, but I suspect the differences between User Mode 3 (Astrophoto mode) and Bulb mode are: -User Mode 3: similar to Av with a certain amount of exposure compensation adequate for night sky, either with wide open aperture or close to it, with white balance set for astrophotography (assuming an area with little to no light pollution from a city), same restrictions as Av concerning shutter speed duration -Bulb Mode: no shutter speed restrictions, every setting of the camera is in your controll Last edited by othar; 12-08-2020 at 06:31 AM. | |
12-08-2020, 07:49 AM | #4 |
Site Supporter |
It's probably a little different in the k70 than my k5 or k50 - which have different firmware setups for the gps-01 as well, but on my k50 you put the camera in bulb mode and then enable the unit in menus and when you select it (with .um .. a back button - I forget which and it's 28 miles away )a little icon turns green, and bulb is replaced by a very long exposure. Make sure you calibrate it in that menu and twisted it all around, BEFORE attaching to the tripod( yep...done that one..doh!). Other tips: Keep your iso very low, usually 200-400 is enough(long explanation available) and set your time about 20% under the cameras suggest max. Set the lens iris a stop or two from bottom to act as a light baffle. This will reduce artifacts and sharpen things up without costing much light. I suggest aiming south this time of year, the winter milkyway is pretty awesome right now. M42 is a great first target, but be ready to deal with the hot center trapezium area in post. Be sure to shoot in raw format so you can stretch shots better. Good luck and hope to see some stars soon. Cheers! Ray |
12-08-2020, 07:55 AM | #5 |
Site Supporter |
The k70 having an "astrophotography" mode sounds interesting.
|
12-08-2020, 08:38 AM - 1 Like | #6 |
Loyal Site Supporter | I would expect that is more like the other scene modes where it is a suggestion to the camera on how to shoot the scene and not something that actually makes use of astrotracer. Much like the night mode or star mode on cellphones. Astro shooting is a lot of trial and error when one is starting out as there is so little light you are outside of any of the program lines for the camera. As you get to know your lenses and shooting spots you will figure out how to run your lenses for best performance. Yes location matters as there light pollution and moon have a huge effect on how your shots turn out. Even things like planets can cause problems if they are in the frame. As far a using the astrotracer it can produce some wonderful results. Most recently I managed this image which is a stack of 94 shots each at 20s, ISO 400, and f/3.5 with my 400mm lens and K-3 using astrotracer and the O-GPS1. So it can produce very good results but it takes practice. My first few times using it didn't go so well even with much shorter lenses. Also for someone starting out invest in a release cable, use the 2s mirror up delay (2s timer) drive mode, and get a rock solid tripod. Since you mentioned you are starting out a great target to shoot now, regardless of what you have for lenses, is Orion. at 35mm to 50mm focal length you can get the majority of the constellation in the frame and with enough shots be able to pull out; Barnard's loop, the witch's hat nebula, the Great Orion Nebula (M42), running man nebula, the horse head nebula, the flame nebula and all sorts of brown dust. Going narrower to into the 100mm to 135mm range you could get nice shots of barnards loop or the witch's hat nebula that come close to filling the frame. At 200mm you would get some great shots of those 2 or you could do a shot of the flame, horse head, running man, and great orion nebula at once. Going up from there you can get detailed shots of 2 of them or with a long enough lens one at a time. There is a lot to shoot in that general area of the sky and the objects there are all pretty bright. |
These users Like MossyRocks's post: |
12-08-2020, 01:32 PM | #7 |
I would expect that is more like the other scene modes where it is a suggestion to the camera on how to shoot the scene and not something that actually makes use of astrotracer. Much like the night mode or star mode on cellphones. Astro shooting is a lot of trial and error when one is starting out as there is so little light you are outside of any of the program lines for the camera. As you get to know your lenses and shooting spots you will figure out how to run your lenses for best performance. Yes location matters as there light pollution and moon have a huge effect on how your shots turn out. Even things like planets can cause problems if they are in the frame. As far a using the astrotracer it can produce some wonderful results. Most recently I managed this image which is a stack of 94 shots each at 20s, ISO 400, and f/3.5 with my 400mm lens and K-3 using astrotracer and the O-GPS1. So it can produce very good results but it takes practice. My first few times using it didn't go so well even with much shorter lenses. Also for someone starting out invest in a release cable, use the 2s mirror up delay (2s timer) drive mode, and get a rock solid tripod. Since you mentioned you are starting out a great target to shoot now, regardless of what you have for lenses, is Orion. at 35mm to 50mm focal length you can get the majority of the constellation in the frame and with enough shots be able to pull out; Barnard's loop, the witch's hat nebula, the Great Orion Nebula (M42), running man nebula, the horse head nebula, the flame nebula and all sorts of brown dust. Going narrower to into the 100mm to 135mm range you could get nice shots of barnards loop or the witch's hat nebula that come close to filling the frame. At 200mm you would get some great shots of those 2 or you could do a shot of the flame, horse head, running man, and great orion nebula at once. Going up from there you can get detailed shots of 2 of them or with a long enough lens one at a time. There is a lot to shoot in that general area of the sky and the objects there are all pretty bright. PENTAX: The choice for astrophotography / explore | RICOH IMAGING | |
12-08-2020, 02:42 PM | #8 |
I only scaned the website quickly so I might have missed it, but where exactly is the difference between User Mode 3 of the K70 and Bulb Mode explained in the article, when used in combination with astrotracer? For me it looked like marketing for the astrotracer, which the OP already bought | |
12-08-2020, 05:22 PM - 2 Likes | #9 |
Loyal Site Supporter | The astrotracer functionality in our cameras functions like an alt-az mount with field rotation making use of the IBIS. To accomplish this the camera has to know how it is oriented in space and where it is located on the surface of earth To accomplish this it makes use of various sensors in the camera and also in the GPS. Involved with this is the electronic compass which is what you are calibrating to attempt to eliminate magnetic variations, the electronic level sensors, and the GPS reciever. The compass tells it which cardinal direction it is facing, the level sensors tell it how it is oriented in relation to the center of the earth, and the GPS tells it where it is located on earth. With this information it can do the math to figure out how to move the sensor to track the sky with reasonable accuracy. This is why things that affect magnetics (power lines, electric motors, rebar, near by vehicles, etc) become so problematic as they throw off the calibration of the magnetic sensor. If you have a poorly calibrated electronic level sensor it will also throw things off as one in my K-3ii is way off and I can't get that thing to track at all unless I have it pointed along the horizon in landscape orientation. Unfortunately there isn't a way that I know of to correct that unless you send it in for service and even then you might get a meh. Apart from your camera's manual the most useful info on astrotracer provided by Ricoh is in the specification area of the O-GPS1 product info page. This actually provides a lot of clues as to what is happening and how it works but requires some understanding of how things appear to moved in the sky when viewed through a camera that has a fixed position for each shot. Lower down ion that page t shows a bunch of different cameras and the theoretical maximum tracking times for a given declination and focal length. One thing that stands out is that regardless of focal length or camera the maximum tracking time at a declination of 90 degrees is always 300s. I will cover declination first in detail so that everyone understands it or hopefully understands it as I have seen lots of people get tripped up with it. A declination of 90 degree does not mean pointed straight up at the zenith but instead means you are pointed at one of the celestial poles. This is the point at which everything in the night sky appears to rotate around. For those in the northern hemisphere this means the spot very near the north star and for those in the southern hemisphere a spot close to the constellation Octans. A declination of 0 means that you are pointed somewhere along the celestial equator. Thing you will find near this area are Orion's belt, the constellation Taurus, and the Pleiades which are all up now. If you are at the north or south pole looking straight up at the zenith the point you are looking at has a declination of 90 degrees and if you look to the horizon you will be pointed at declination 0. Now if you are on the equator and look straight up at the zenith you will be looking some where along the celestial equator with a declination of 0 but looking either true north to the horizon or to true south to the horizon means you will be pointing at a declination of 90 degrees. If you are some where in between the equator and pole the celestial pole (declination of 90 degrees) will be somewhere between the horizon and the zenith along the meridian that runs through the spot where your tripod is setup. Now that everyone should understand what declination means we can go back to the theoretical maximum 300s tracking regardless of camera or focal length when point at a declination of 90 degrees. If we are pointed exactly at the celestial pole (declination of 90 degrees) in the middle of the frame and have a perfect calibration and our camera correctly figures out that it is actually point at the celestial pole with this means that in the frame all of the motion will be rotational. So to accurately track the night sky the camera only has to rotate the sensor. The rotation of the sensor has to happen at the rate of 15 degrees per hour since this is the amount that the earth rotates and that is what astrotracer has to compensate for. If I remember correctly the sensors in these cameras are capable of something close to 1 degree of rotation. So the sensor can accurately track around the celestial pole for 1/15 hours or 240s with that assumption. Now there might be a bit more tracking or maybe there is the ability to back up the sensor some in rotation, I don't know, so the 300s quote that Ricoh is making seems reasonable especially if you assume acceptable out of round stars instead of perfect points. Focal length doesn't matter here because everything is rotating about the center of the frame at the same angular speed and there is no linear component that needs compensating for. Now if instead we point at something along the celestial equator (declination of 0 degrees) using a normal or long lens, chosen because it simplifies things for now but later I will go into wides and ultra wides as strange things happen with them, we would find that all the movement across the frame is basically linear but that things are really moving across our frame. There is no rotational part as we are as far from the celestial pole as we can be and if you have ever seen an equatorial mount working the axis that the counter weight is on will trace this nice straight line through the sky for you. For the camera to accurately track this it is limited by the linear x/y movement that the camera can do. For my K-3 I think this is 1.5mm but the value isn't important only that the less travel the sensor has the shorter the theoretical tracking is. Here focal length does matter as using a lens 2x as long means that the stars travel across a sensor twice as fast and thus can only be tracked for half the time. This area of the sky is also where we see the shortest theoretical tracking times which does make sense since there is such a limited amount of travel for the sensor. Now if we pick some spot that has a declination between 0 and 90 degrees and point the camera there we will find that there will be some rotational component to the movement of the stars across the frame and some linear component of their movement across the frame. Here the sensor will have to do both a x/y translation of the sensor and a rotation of it to track the night sky correctly. The amount of each varies but there will be more linear movement as you are pointed more towards the celestial equator (declination 0) and more rotational movement as you are pointed more towards the celestial pole (declination 90). Because of this your theoretical maximum tracking time will be some value between the times for tracking when pointing at the celestial pole and pointing at the celestial equator and focal length does matter. Finally we can look at using wides and ultra wide lenses and some of the strangeness that happens with them. These will produce the best results when they are pointed either at the celestial pole. This is because there is only rotation and everything is rotating at the same angular speed, the stars in the center of the frame, the stars at the edge of the frame, the center of the sensor, and the edge of the sensor. So you can get those nice 5 minute exposures with nice stars without issue. However as you move away from the celestial equator you now have introduced a linear component to the movement of the stars. The rotation of the sensor only happens around the center of the sensor so now you will start to get some trailing out at the edges from errors. If we point a wide or ultrawide along the celestial equator the movement of the sensor will only be linear but because of the large field of view stars at the edge of the frame will have rotational components, and the wider the lens the greater the rotational component will be for the star farther away from the center of the frame. In looking at the shots in the link you provided we do indeed start to see these artefacts in the 60 and 90 second shots with focal lengths in the teens. Now with these wide and ultra wide lenses (lets say a field of view greater than that of a 28mm lens on full frame) we get into some interesting aspects. If I use my 12mm lens I can get really nice pinpoint stars without astrotracer doing 20s exposures and do 30 second ones if I accept starts that are a bit out of round. If I were to use astrotracer with my 12mm the sensor would actually move very slowly linearly and provided I have it pointed away from the celestial equator so it really isn't in the shot I would also accept some slightly trailed stars at the edges and corners because of the rotation of the sensor. If I did a 40s shot with that lens I would expect very good results with corners probably to similar to that of a 30s untracked shot and moving up to 60 seconds I would expect to start to see trailing in at least one corner but maybe more depending on where I had it pointed. From there it becomes how much trailing am I willing to accept as it will increase in length and move farther into the frame. In the link the longest shot with one of those really wide lenses was 90s and is actually pretty good but it is clearly showing trailing in all the corners which is predicted behavior. |
These users Like MossyRocks's post: |
12-08-2020, 05:22 PM | #10 |
I only scaned the website quickly so I might have missed it, but where exactly is the difference between User Mode 3 of the K70 and Bulb Mode explained in the article, when used in combination with astrotracer? For me it looked like marketing for the astrotracer, which the OP already bought Last edited by SharkyCA; 12-08-2020 at 05:28 PM. | |
12-08-2020, 05:38 PM | #11 |
That link was rather light on actual details and mostly a marketing piece. While it was a good intro to using astrotracer and sets some expectations it doesn't provide the necessary info to really master it. The O-GPS1 unit is a wonderful little device and can actually be pushed much harder than your link indicates as they were rather insistent of it only being useful out to about 135mm, yet the image I linked to before (the great orion nebula) was done using astrotracer with the O-GPS1 on my K-3 and my 400mm lens. That long focal length really pushes astrotracer as it will show any calibration issues and at 400mm there are almost always some which is why I use 20s shots instead of the theoretical 60ish the specs would indicate I could use. Additionally the Great Orion Nebula has a declination of almost 0 being slightly south of the celestial equator so has a very high linear speed across the sensor which also taxes astrotracer. Also after reading that link it doesn't make any mention of the Night Scene mode being used. I went and looked in the K-70 manual and it is very light on details of what the camera does for Night Scene but in the astrotracer section it basically reads like the same as it does for my K-3 stating that bulb mode is what it needs to be put in. The astrotracer functionality in our cameras functions like an alt-az mount with field rotation making use of the IBIS. To accomplish this the camera has to know how it is oriented in space and where it is located on the surface of earth To accomplish this it makes use of various sensors in the camera and also in the GPS. Involved with this is the electronic compass which is what you are calibrating to attempt to eliminate magnetic variations, the electronic level sensors, and the GPS reciever. The compass tells it which cardinal direction it is facing, the level sensors tell it how it is oriented in relation to the center of the earth, and the GPS tells it where it is located on earth. With this information it can do the math to figure out how to move the sensor to track the sky with reasonable accuracy. This is why things that affect magnetics (power lines, electric motors, rebar, near by vehicles, etc) become so problematic as they throw off the calibration of the magnetic sensor. If you have a poorly calibrated electronic level sensor it will also throw things off as one in my K-3ii is way off and I can't get that thing to track at all unless I have it pointed along the horizon in landscape orientation. Unfortunately there isn't a way that I know of to correct that unless you send it in for service and even then you might get a meh. Apart from your camera's manual the most useful info on astrotracer provided by Ricoh is in the specification area of the O-GPS1 product info page. This actually provides a lot of clues as to what is happening and how it works but requires some understanding of how things appear to moved in the sky when viewed through a camera that has a fixed position for each shot. Lower down ion that page t shows a bunch of different cameras and the theoretical maximum tracking times for a given declination and focal length. One thing that stands out is that regardless of focal length or camera the maximum tracking time at a declination of 90 degrees is always 300s. I will cover declination first in detail so that everyone understands it or hopefully understands it as I have seen lots of people get tripped up with it. A declination of 90 degree does not mean pointed straight up at the zenith but instead means you are pointed at one of the celestial poles. This is the point at which everything in the night sky appears to rotate around. For those in the northern hemisphere this means the spot very near the north star and for those in the southern hemisphere a spot close to the constellation Octans. A declination of 0 means that you are pointed somewhere along the celestial equator. Thing you will find near this area are Orion's belt, the constellation Taurus, and the Pleiades which are all up now. If you are at the north or south pole looking straight up at the zenith the point you are looking at has a declination of 90 degrees and if you look to the horizon you will be pointed at declination 0. Now if you are on the equator and look straight up at the zenith you will be looking some where along the celestial equator with a declination of 0 but looking either true north to the horizon or to true south to the horizon means you will be pointing at a declination of 90 degrees. If you are some where in between the equator and pole the celestial pole (declination of 90 degrees) will be somewhere between the horizon and the zenith along the meridian that runs through the spot where your tripod is setup. Now that everyone should understand what declination means we can go back to the theoretical maximum 300s tracking regardless of camera or focal length when point at a declination of 90 degrees. If we are pointed exactly at the celestial pole (declination of 90 degrees) in the middle of the frame and have a perfect calibration and our camera correctly figures out that it is actually point at the celestial pole with this means that in the frame all of the motion will be rotational. So to accurately track the night sky the camera only has to rotate the sensor. The rotation of the sensor has to happen at the rate of 15 degrees per hour since this is the amount that the earth rotates and that is what astrotracer has to compensate for. If I remember correctly the sensors in these cameras are capable of something close to 1 degree of rotation. So the sensor can accurately track around the celestial pole for 1/15 hours or 240s with that assumption. Now there might be a bit more tracking or maybe there is the ability to back up the sensor some in rotation, I don't know, so the 300s quote that Ricoh is making seems reasonable especially if you assume acceptable out of round stars instead of perfect points. Focal length doesn't matter here because everything is rotating about the center of the frame at the same angular speed and there is no linear component that needs compensating for. Now if instead we point at something along the celestial equator (declination of 0 degrees) using a normal or long lens, chosen because it simplifies things for now but later I will go into wides and ultra wides as strange things happen with them, we would find that all the movement across the frame is basically linear but that things are really moving across our frame. There is no rotational part as we are as far from the celestial pole as we can be and if you have ever seen an equatorial mount working the axis that the counter weight is on will trace this nice straight line through the sky for you. For the camera to accurately track this it is limited by the linear x/y movement that the camera can do. For my K-3 I think this is 1.5mm but the value isn't important only that the less travel the sensor has the shorter the theoretical tracking is. Here focal length does matter as using a lens 2x as long means that the stars travel across a sensor twice as fast and thus can only be tracked for half the time. This area of the sky is also where we see the shortest theoretical tracking times which does make sense since there is such a limited amount of travel for the sensor. Now if we pick some spot that has a declination between 0 and 90 degrees and point the camera there we will find that there will be some rotational component to the movement of the stars across the frame and some linear component of their movement across the frame. Here the sensor will have to do both a x/y translation of the sensor and a rotation of it to track the night sky correctly. The amount of each varies but there will be more linear movement as you are pointed more towards the celestial equator (declination 0) and more rotational movement as you are pointed more towards the celestial pole (declination 90). Because of this your theoretical maximum tracking time will be some value between the times for tracking when pointing at the celestial pole and pointing at the celestial equator and focal length does matter. Finally we can look at using wides and ultra wide lenses and some of the strangeness that happens with them. These will produce the best results when they are pointed either at the celestial pole. This is because there is only rotation and everything is rotating at the same angular speed, the stars in the center of the frame, the stars at the edge of the frame, the center of the sensor, and the edge of the sensor. So you can get those nice 5 minute exposures with nice stars without issue. However as you move away from the celestial equator you now have introduced a linear component to the movement of the stars. The rotation of the sensor only happens around the center of the sensor so now you will start to get some trailing out at the edges from errors. If we point a wide or ultrawide along the celestial equator the movement of the sensor will only be linear but because of the large field of view stars at the edge of the frame will have rotational components, and the wider the lens the greater the rotational component will be for the star farther away from the center of the frame. In looking at the shots in the link you provided we do indeed start to see these artefacts in the 60 and 90 second shots with focal lengths in the teens. Now with these wide and ultra wide lenses (lets say a field of view greater than that of a 28mm lens on full frame) we get into some interesting aspects. If I use my 12mm lens I can get really nice pinpoint stars without astrotracer doing 20s exposures and do 30 second ones if I accept starts that are a bit out of round. If I were to use astrotracer with my 12mm the sensor would actually move very slowly linearly and provided I have it pointed away from the celestial equator so it really isn't in the shot I would also accept some slightly trailed stars at the edges and corners because of the rotation of the sensor. If I did a 40s shot with that lens I would expect very good results with corners probably to similar to that of a 30s untracked shot and moving up to 60 seconds I would expect to start to see trailing in at least one corner but maybe more depending on where I had it pointed. From there it becomes how much trailing am I willing to accept as it will increase in length and move farther into the frame. In the link the longest shot with one of those really wide lenses was 90s and is actually pretty good but it is clearly showing trailing in all the corners which is predicted behavior. Cheers | |
12-08-2020, 08:10 PM | #12 |
Hi, Welcome to the forums! I don't do Astro myself, but the following link might be of some use! PENTAX: The choice for astrophotography / explore | RICOH IMAGING Cheers! ---------- Post added 12-08-20 at 08:10 PM ---------- I don't own a K-70 or a GPS unit so I am merely guessing based on the names, but I suspect the differences between User Mode 3 (Astrophoto mode) and Bulb mode are: -User Mode 3: similar to Av with a certain amount of exposure compensation adequate for night sky, either with wide open aperture or close to it, with white balance set for astrophotography (assuming an area with little to no light pollution from a city), same restrictions as Av concerning shutter speed duration -Bulb Mode: no shutter speed restrictions, every setting of the camera is in your controll ---------- Post added 12-08-20 at 08:12 PM ---------- It's probably a little different in the k70 than my k5 or k50 - which have different firmware setups for the gps-01 as well, but on my k50 you put the camera in bulb mode and then enable the unit in menus and when you select it (with .um .. a back button - I forget which and it's 28 miles away )a little icon turns green, and bulb is replaced by a very long exposure. Make sure you calibrate it in that menu and twisted it all around, BEFORE attaching to the tripod( yep...done that one..doh!). Other tips: Keep your iso very low, usually 200-400 is enough(long explanation available) and set your time about 20% under the cameras suggest max. Set the lens iris a stop or two from bottom to act as a light baffle. This will reduce artifacts and sharpen things up without costing much light. I suggest aiming south this time of year, the winter milkyway is pretty awesome right now. M42 is a great first target, but be ready to deal with the hot center trapezium area in post. Be sure to shoot in raw format so you can stretch shots better. Good luck and hope to see some stars soon. Cheers! Ray ---------- Post added 12-08-20 at 08:13 PM ---------- That link was rather light on actual details and mostly a marketing piece. While it was a good intro to using astrotracer and sets some expectations it doesn't provide the necessary info to really master it. The O-GPS1 unit is a wonderful little device and can actually be pushed much harder than your link indicates as they were rather insistent of it only being useful out to about 135mm, yet the image I linked to before (the great orion nebula) was done using astrotracer with the O-GPS1 on my K-3 and my 400mm lens. That long focal length really pushes astrotracer as it will show any calibration issues and at 400mm there are almost always some which is why I use 20s shots instead of the theoretical 60ish the specs would indicate I could use. Additionally the Great Orion Nebula has a declination of almost 0 being slightly south of the celestial equator so has a very high linear speed across the sensor which also taxes astrotracer. Also after reading that link it doesn't make any mention of the Night Scene mode being used. I went and looked in the K-70 manual and it is very light on details of what the camera does for Night Scene but in the astrotracer section it basically reads like the same as it does for my K-3 stating that bulb mode is what it needs to be put in. The astrotracer functionality in our cameras functions like an alt-az mount with field rotation making use of the IBIS. To accomplish this the camera has to know how it is oriented in space and where it is located on the surface of earth To accomplish this it makes use of various sensors in the camera and also in the GPS. Involved with this is the electronic compass which is what you are calibrating to attempt to eliminate magnetic variations, the electronic level sensors, and the GPS reciever. The compass tells it which cardinal direction it is facing, the level sensors tell it how it is oriented in relation to the center of the earth, and the GPS tells it where it is located on earth. With this information it can do the math to figure out how to move the sensor to track the sky with reasonable accuracy. This is why things that affect magnetics (power lines, electric motors, rebar, near by vehicles, etc) become so problematic as they throw off the calibration of the magnetic sensor. If you have a poorly calibrated electronic level sensor it will also throw things off as one in my K-3ii is way off and I can't get that thing to track at all unless I have it pointed along the horizon in landscape orientation. Unfortunately there isn't a way that I know of to correct that unless you send it in for service and even then you might get a meh. Apart from your camera's manual the most useful info on astrotracer provided by Ricoh is in the specification area of the O-GPS1 product info page. This actually provides a lot of clues as to what is happening and how it works but requires some understanding of how things appear to moved in the sky when viewed through a camera that has a fixed position for each shot. Lower down ion that page t shows a bunch of different cameras and the theoretical maximum tracking times for a given declination and focal length. One thing that stands out is that regardless of focal length or camera the maximum tracking time at a declination of 90 degrees is always 300s. I will cover declination first in detail so that everyone understands it or hopefully understands it as I have seen lots of people get tripped up with it. A declination of 90 degree does not mean pointed straight up at the zenith but instead means you are pointed at one of the celestial poles. This is the point at which everything in the night sky appears to rotate around. For those in the northern hemisphere this means the spot very near the north star and for those in the southern hemisphere a spot close to the constellation Octans. A declination of 0 means that you are pointed somewhere along the celestial equator. Thing you will find near this area are Orion's belt, the constellation Taurus, and the Pleiades which are all up now. If you are at the north or south pole looking straight up at the zenith the point you are looking at has a declination of 90 degrees and if you look to the horizon you will be pointed at declination 0. Now if you are on the equator and look straight up at the zenith you will be looking some where along the celestial equator with a declination of 0 but looking either true north to the horizon or to true south to the horizon means you will be pointing at a declination of 90 degrees. If you are some where in between the equator and pole the celestial pole (declination of 90 degrees) will be somewhere between the horizon and the zenith along the meridian that runs through the spot where your tripod is setup. Now that everyone should understand what declination means we can go back to the theoretical maximum 300s tracking regardless of camera or focal length when point at a declination of 90 degrees. If we are pointed exactly at the celestial pole (declination of 90 degrees) in the middle of the frame and have a perfect calibration and our camera correctly figures out that it is actually point at the celestial pole with this means that in the frame all of the motion will be rotational. So to accurately track the night sky the camera only has to rotate the sensor. The rotation of the sensor has to happen at the rate of 15 degrees per hour since this is the amount that the earth rotates and that is what astrotracer has to compensate for. If I remember correctly the sensors in these cameras are capable of something close to 1 degree of rotation. So the sensor can accurately track around the celestial pole for 1/15 hours or 240s with that assumption. Now there might be a bit more tracking or maybe there is the ability to back up the sensor some in rotation, I don't know, so the 300s quote that Ricoh is making seems reasonable especially if you assume acceptable out of round stars instead of perfect points. Focal length doesn't matter here because everything is rotating about the center of the frame at the same angular speed and there is no linear component that needs compensating for. Now if instead we point at something along the celestial equator (declination of 0 degrees) using a normal or long lens, chosen because it simplifies things for now but later I will go into wides and ultra wides as strange things happen with them, we would find that all the movement across the frame is basically linear but that things are really moving across our frame. There is no rotational part as we are as far from the celestial pole as we can be and if you have ever seen an equatorial mount working the axis that the counter weight is on will trace this nice straight line through the sky for you. For the camera to accurately track this it is limited by the linear x/y movement that the camera can do. For my K-3 I think this is 1.5mm but the value isn't important only that the less travel the sensor has the shorter the theoretical tracking is. Here focal length does matter as using a lens 2x as long means that the stars travel across a sensor twice as fast and thus can only be tracked for half the time. This area of the sky is also where we see the shortest theoretical tracking times which does make sense since there is such a limited amount of travel for the sensor. Now if we pick some spot that has a declination between 0 and 90 degrees and point the camera there we will find that there will be some rotational component to the movement of the stars across the frame and some linear component of their movement across the frame. Here the sensor will have to do both a x/y translation of the sensor and a rotation of it to track the night sky correctly. The amount of each varies but there will be more linear movement as you are pointed more towards the celestial equator (declination 0) and more rotational movement as you are pointed more towards the celestial pole (declination 90). Because of this your theoretical maximum tracking time will be some value between the times for tracking when pointing at the celestial pole and pointing at the celestial equator and focal length does matter. Finally we can look at using wides and ultra wide lenses and some of the strangeness that happens with them. These will produce the best results when they are pointed either at the celestial pole. This is because there is only rotation and everything is rotating at the same angular speed, the stars in the center of the frame, the stars at the edge of the frame, the center of the sensor, and the edge of the sensor. So you can get those nice 5 minute exposures with nice stars without issue. However as you move away from the celestial equator you now have introduced a linear component to the movement of the stars. The rotation of the sensor only happens around the center of the sensor so now you will start to get some trailing out at the edges from errors. If we point a wide or ultrawide along the celestial equator the movement of the sensor will only be linear but because of the large field of view stars at the edge of the frame will have rotational components, and the wider the lens the greater the rotational component will be for the star farther away from the center of the frame. In looking at the shots in the link you provided we do indeed start to see these artefacts in the 60 and 90 second shots with focal lengths in the teens. Now with these wide and ultra wide lenses (lets say a field of view greater than that of a 28mm lens on full frame) we get into some interesting aspects. If I use my 12mm lens I can get really nice pinpoint stars without astrotracer doing 20s exposures and do 30 second ones if I accept starts that are a bit out of round. If I were to use astrotracer with my 12mm the sensor would actually move very slowly linearly and provided I have it pointed away from the celestial equator so it really isn't in the shot I would also accept some slightly trailed stars at the edges and corners because of the rotation of the sensor. If I did a 40s shot with that lens I would expect very good results with corners probably to similar to that of a 30s untracked shot and moving up to 60 seconds I would expect to start to see trailing in at least one corner but maybe more depending on where I had it pointed. From there it becomes how much trailing am I willing to accept as it will increase in length and move farther into the frame. In the link the longest shot with one of those really wide lenses was 90s and is actually pretty good but it is clearly showing trailing in all the corners which is predicted behavior. | |
12-08-2020, 08:13 PM | #13 |
12-08-2020, 08:15 PM - 2 Likes | #14 |
That link was rather light on actual details and mostly a marketing piece. While it was a good intro to using astrotracer and sets some expectations it doesn't provide the necessary info to really master it. The O-GPS1 unit is a wonderful little device and can actually be pushed much harder than your link indicates as they were rather insistent of it only being useful out to about 135mm, yet the image I linked to before (the great orion nebula) was done using astrotracer with the O-GPS1 on my K-3 and my 400mm lens. That long focal length really pushes astrotracer as it will show any calibration issues and at 400mm there are almost always some which is why I use 20s shots instead of the theoretical 60ish the specs would indicate I could use. Additionally the Great Orion Nebula has a declination of almost 0 being slightly south of the celestial equator so has a very high linear speed across the sensor which also taxes astrotracer. Also after reading that link it doesn't make any mention of the Night Scene mode being used. I went and looked in the K-70 manual and it is very light on details of what the camera does for Night Scene but in the astrotracer section it basically reads like the same as it does for my K-3 stating that bulb mode is what it needs to be put in. The astrotracer functionality in our cameras functions like an alt-az mount with field rotation making use of the IBIS. To accomplish this the camera has to know how it is oriented in space and where it is located on the surface of earth To accomplish this it makes use of various sensors in the camera and also in the GPS. Involved with this is the electronic compass which is what you are calibrating to attempt to eliminate magnetic variations, the electronic level sensors, and the GPS reciever. The compass tells it which cardinal direction it is facing, the level sensors tell it how it is oriented in relation to the center of the earth, and the GPS tells it where it is located on earth. With this information it can do the math to figure out how to move the sensor to track the sky with reasonable accuracy. This is why things that affect magnetics (power lines, electric motors, rebar, near by vehicles, etc) become so problematic as they throw off the calibration of the magnetic sensor. If you have a poorly calibrated electronic level sensor it will also throw things off as one in my K-3ii is way off and I can't get that thing to track at all unless I have it pointed along the horizon in landscape orientation. Unfortunately there isn't a way that I know of to correct that unless you send it in for service and even then you might get a meh. Apart from your camera's manual the most useful info on astrotracer provided by Ricoh is in the specification area of the O-GPS1 product info page. This actually provides a lot of clues as to what is happening and how it works but requires some understanding of how things appear to moved in the sky when viewed through a camera that has a fixed position for each shot. Lower down ion that page t shows a bunch of different cameras and the theoretical maximum tracking times for a given declination and focal length. One thing that stands out is that regardless of focal length or camera the maximum tracking time at a declination of 90 degrees is always 300s. I will cover declination first in detail so that everyone understands it or hopefully understands it as I have seen lots of people get tripped up with it. A declination of 90 degree does not mean pointed straight up at the zenith but instead means you are pointed at one of the celestial poles. This is the point at which everything in the night sky appears to rotate around. For those in the northern hemisphere this means the spot very near the north star and for those in the southern hemisphere a spot close to the constellation Octans. A declination of 0 means that you are pointed somewhere along the celestial equator. Thing you will find near this area are Orion's belt, the constellation Taurus, and the Pleiades which are all up now. If you are at the north or south pole looking straight up at the zenith the point you are looking at has a declination of 90 degrees and if you look to the horizon you will be pointed at declination 0. Now if you are on the equator and look straight up at the zenith you will be looking some where along the celestial equator with a declination of 0 but looking either true north to the horizon or to true south to the horizon means you will be pointing at a declination of 90 degrees. If you are some where in between the equator and pole the celestial pole (declination of 90 degrees) will be somewhere between the horizon and the zenith along the meridian that runs through the spot where your tripod is setup. Now that everyone should understand what declination means we can go back to the theoretical maximum 300s tracking regardless of camera or focal length when point at a declination of 90 degrees. If we are pointed exactly at the celestial pole (declination of 90 degrees) in the middle of the frame and have a perfect calibration and our camera correctly figures out that it is actually point at the celestial pole with this means that in the frame all of the motion will be rotational. So to accurately track the night sky the camera only has to rotate the sensor. The rotation of the sensor has to happen at the rate of 15 degrees per hour since this is the amount that the earth rotates and that is what astrotracer has to compensate for. If I remember correctly the sensors in these cameras are capable of something close to 1 degree of rotation. So the sensor can accurately track around the celestial pole for 1/15 hours or 240s with that assumption. Now there might be a bit more tracking or maybe there is the ability to back up the sensor some in rotation, I don't know, so the 300s quote that Ricoh is making seems reasonable especially if you assume acceptable out of round stars instead of perfect points. Focal length doesn't matter here because everything is rotating about the center of the frame at the same angular speed and there is no linear component that needs compensating for. Now if instead we point at something along the celestial equator (declination of 0 degrees) using a normal or long lens, chosen because it simplifies things for now but later I will go into wides and ultra wides as strange things happen with them, we would find that all the movement across the frame is basically linear but that things are really moving across our frame. There is no rotational part as we are as far from the celestial pole as we can be and if you have ever seen an equatorial mount working the axis that the counter weight is on will trace this nice straight line through the sky for you. For the camera to accurately track this it is limited by the linear x/y movement that the camera can do. For my K-3 I think this is 1.5mm but the value isn't important only that the less travel the sensor has the shorter the theoretical tracking is. Here focal length does matter as using a lens 2x as long means that the stars travel across a sensor twice as fast and thus can only be tracked for half the time. This area of the sky is also where we see the shortest theoretical tracking times which does make sense since there is such a limited amount of travel for the sensor. Now if we pick some spot that has a declination between 0 and 90 degrees and point the camera there we will find that there will be some rotational component to the movement of the stars across the frame and some linear component of their movement across the frame. Here the sensor will have to do both a x/y translation of the sensor and a rotation of it to track the night sky correctly. The amount of each varies but there will be more linear movement as you are pointed more towards the celestial equator (declination 0) and more rotational movement as you are pointed more towards the celestial pole (declination 90). Because of this your theoretical maximum tracking time will be some value between the times for tracking when pointing at the celestial pole and pointing at the celestial equator and focal length does matter. Finally we can look at using wides and ultra wide lenses and some of the strangeness that happens with them. These will produce the best results when they are pointed either at the celestial pole. This is because there is only rotation and everything is rotating at the same angular speed, the stars in the center of the frame, the stars at the edge of the frame, the center of the sensor, and the edge of the sensor. So you can get those nice 5 minute exposures with nice stars without issue. However as you move away from the celestial equator you now have introduced a linear component to the movement of the stars. The rotation of the sensor only happens around the center of the sensor so now you will start to get some trailing out at the edges from errors. If we point a wide or ultrawide along the celestial equator the movement of the sensor will only be linear but because of the large field of view stars at the edge of the frame will have rotational components, and the wider the lens the greater the rotational component will be for the star farther away from the center of the frame. In looking at the shots in the link you provided we do indeed start to see these artefacts in the 60 and 90 second shots with focal lengths in the teens. Now with these wide and ultra wide lenses (lets say a field of view greater than that of a 28mm lens on full frame) we get into some interesting aspects. If I use my 12mm lens I can get really nice pinpoint stars without astrotracer doing 20s exposures and do 30 second ones if I accept starts that are a bit out of round. If I were to use astrotracer with my 12mm the sensor would actually move very slowly linearly and provided I have it pointed away from the celestial equator so it really isn't in the shot I would also accept some slightly trailed stars at the edges and corners because of the rotation of the sensor. If I did a 40s shot with that lens I would expect very good results with corners probably to similar to that of a 30s untracked shot and moving up to 60 seconds I would expect to start to see trailing in at least one corner but maybe more depending on where I had it pointed. From there it becomes how much trailing am I willing to accept as it will increase in length and move farther into the frame. In the link the longest shot with one of those really wide lenses was 90s and is actually pretty good but it is clearly showing trailing in all the corners which is predicted behavior. There is some great info there. Your help is very much appreciated. In the end, I will try both Astrophoto and Bulb modes with the Astro tracer and see what happens. | |
These users Like Simmo25's post: |
12-09-2020, 06:33 AM - 1 Like | #15 |
Site Supporter |
I'll second the need for a good solid tripod and remote. An intervalometers is also inexpensive and very handy, for pretty much any situation where you want a bunch of shots --time lapse, stacking etc. even daytime stuff sometimes. Good luck! Can't wait to see some results, and there's an astronomy/ap group here too if you need more guidance! |
These users Like blues_hawk's post: |
|
Bookmarks |
Tags - Make this thread easier to find by adding keywords to it! |
astrophoto, astrotracer, av, battery, bulb, camera, degrees, exposure, frame, k-70, k50, k70, lens, light, mode, movement, pentax k-70, pole, post, restrictions, sensor, shutter, sky, stars, telescope, thanks, time |
Thread Tools | Search this Thread |
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Astrotracer-Bulb mode...long exposure | OSRags | Pentax K-1 & K-1 II | 2 | 05-01-2019 09:09 PM |
Pentax K-70 can't get astrotracer to work | Landy | Troubleshooting and Beginner Help | 3 | 09-23-2018 08:52 AM |
Proper use of Astrotracer? | awscreo | Pentax DSLR Discussion | 13 | 07-29-2017 12:22 PM |
Which Support for K3II Astrotracer or Pixel Shift Feature Use | C_Jones | Pentax Camera and Field Accessories | 33 | 11-25-2015 12:40 PM |
Do you see use for an astrotracer either from ogps 1 or K3-II ? | Nicolas06 | Pentax SLR Lens Discussion | 20 | 05-24-2015 02:15 PM |